| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.633204633204633, | |
| "eval_steps": 200, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15444015444015444, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 188.4417, | |
| "eval_samples_per_second": 37.264, | |
| "eval_steps_per_second": 4.659, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3088803088803089, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 187.0753, | |
| "eval_samples_per_second": 37.536, | |
| "eval_steps_per_second": 4.693, | |
| "eval_wer": 0.8660615740199706, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3861003861003861, | |
| "grad_norm": 2.7789804935455322, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 3.7305, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.46332046332046334, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 186.9186, | |
| "eval_samples_per_second": 37.567, | |
| "eval_steps_per_second": 4.697, | |
| "eval_wer": 0.7040110111279914, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6177606177606177, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 187.3522, | |
| "eval_samples_per_second": 37.48, | |
| "eval_steps_per_second": 4.686, | |
| "eval_wer": 0.5505953540311376, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7722007722007722, | |
| "grad_norm": 11.684309959411621, | |
| "learning_rate": 0.0002731090909090909, | |
| "loss": 0.8464, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7722007722007722, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 187.5603, | |
| "eval_samples_per_second": 37.439, | |
| "eval_steps_per_second": 4.681, | |
| "eval_wer": 0.5168088504538195, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9266409266409267, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 187.4249, | |
| "eval_samples_per_second": 37.466, | |
| "eval_steps_per_second": 4.685, | |
| "eval_wer": 0.4824769844052303, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 188.7764, | |
| "eval_samples_per_second": 37.197, | |
| "eval_steps_per_second": 4.651, | |
| "eval_wer": 0.46011712308311586, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.1583011583011582, | |
| "grad_norm": 0.6005635261535645, | |
| "learning_rate": 0.0002458363636363636, | |
| "loss": 0.6629, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.2355212355212355, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 188.0077, | |
| "eval_samples_per_second": 37.35, | |
| "eval_steps_per_second": 4.67, | |
| "eval_wer": 0.4445483230104008, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.3899613899613898, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 188.204, | |
| "eval_samples_per_second": 37.311, | |
| "eval_steps_per_second": 4.665, | |
| "eval_wer": 0.41425473621336656, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.5444015444015444, | |
| "grad_norm": 0.4440418481826782, | |
| "learning_rate": 0.0002186181818181818, | |
| "loss": 0.5655, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.5444015444015444, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 188.2865, | |
| "eval_samples_per_second": 37.294, | |
| "eval_steps_per_second": 4.663, | |
| "eval_wer": 0.41700751821121107, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.698841698841699, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.1618, | |
| "eval_samples_per_second": 37.122, | |
| "eval_steps_per_second": 4.642, | |
| "eval_wer": 0.4047108929661226, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.8532818532818531, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 188.2544, | |
| "eval_samples_per_second": 37.301, | |
| "eval_steps_per_second": 4.664, | |
| "eval_wer": 0.3966213496422682, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.9305019305019306, | |
| "grad_norm": 0.9740249514579773, | |
| "learning_rate": 0.00019150909090909088, | |
| "loss": 0.5524, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0077220077220077, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.1265, | |
| "eval_samples_per_second": 37.129, | |
| "eval_steps_per_second": 4.642, | |
| "eval_wer": 0.37794917741160583, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 188.7855, | |
| "eval_samples_per_second": 37.196, | |
| "eval_steps_per_second": 4.651, | |
| "eval_wer": 0.37366418656590444, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.3166023166023164, | |
| "grad_norm": 0.5873022675514221, | |
| "learning_rate": 0.00016429090909090907, | |
| "loss": 0.4773, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.3166023166023164, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.7378, | |
| "eval_samples_per_second": 37.009, | |
| "eval_steps_per_second": 4.627, | |
| "eval_wer": 0.3698336644462623, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.471042471042471, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.8812, | |
| "eval_samples_per_second": 36.981, | |
| "eval_steps_per_second": 4.624, | |
| "eval_wer": 0.37235271967070493, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.6254826254826256, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.4769, | |
| "eval_samples_per_second": 37.06, | |
| "eval_steps_per_second": 4.634, | |
| "eval_wer": 0.3583940373703141, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 1.4725390672683716, | |
| "learning_rate": 0.00013712727272727272, | |
| "loss": 0.4694, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.7799227799227797, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.6126, | |
| "eval_samples_per_second": 37.033, | |
| "eval_steps_per_second": 4.63, | |
| "eval_wer": 0.3820783504083726, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.9343629343629343, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 190.0131, | |
| "eval_samples_per_second": 36.955, | |
| "eval_steps_per_second": 4.621, | |
| "eval_wer": 0.4729850804409645, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.088803088803089, | |
| "grad_norm": 14.039852142333984, | |
| "learning_rate": 0.0001099090909090909, | |
| "loss": 0.6537, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.088803088803089, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.7145, | |
| "eval_samples_per_second": 37.014, | |
| "eval_steps_per_second": 4.628, | |
| "eval_wer": 0.4753613026372171, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.2432432432432434, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.6267, | |
| "eval_samples_per_second": 37.031, | |
| "eval_steps_per_second": 4.63, | |
| "eval_wer": 0.5899263760663784, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 3.3976833976833976, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.2072, | |
| "eval_samples_per_second": 37.113, | |
| "eval_steps_per_second": 4.64, | |
| "eval_wer": 0.5957565605806812, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 3.474903474903475, | |
| "grad_norm": 15.850507736206055, | |
| "learning_rate": 8.269090909090907e-05, | |
| "loss": 0.8238, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.552123552123552, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 188.9868, | |
| "eval_samples_per_second": 37.156, | |
| "eval_steps_per_second": 4.646, | |
| "eval_wer": 0.633646267513277, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 3.7065637065637067, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 190.0461, | |
| "eval_samples_per_second": 36.949, | |
| "eval_steps_per_second": 4.62, | |
| "eval_wer": 0.6025865762923143, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 3.861003861003861, | |
| "grad_norm": 6.711777687072754, | |
| "learning_rate": 5.547272727272727e-05, | |
| "loss": 0.8682, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.861003861003861, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.1358, | |
| "eval_samples_per_second": 37.127, | |
| "eval_steps_per_second": 4.642, | |
| "eval_wer": 0.5671380156596938, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.015444015444015, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 190.8009, | |
| "eval_samples_per_second": 36.803, | |
| "eval_steps_per_second": 4.602, | |
| "eval_wer": 0.5377923207770117, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 4.1698841698841695, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.3706, | |
| "eval_samples_per_second": 37.081, | |
| "eval_steps_per_second": 4.636, | |
| "eval_wer": 0.5373638216924416, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 4.2471042471042475, | |
| "grad_norm": 6.047791957855225, | |
| "learning_rate": 2.8309090909090903e-05, | |
| "loss": 0.855, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.324324324324325, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 189.8451, | |
| "eval_samples_per_second": 36.988, | |
| "eval_steps_per_second": 4.625, | |
| "eval_wer": 0.5328451040733383, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 4.478764478764479, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 190.1917, | |
| "eval_samples_per_second": 36.921, | |
| "eval_steps_per_second": 4.616, | |
| "eval_wer": 0.5225351564021659, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 4.633204633204633, | |
| "grad_norm": 9.171218872070312, | |
| "learning_rate": 1.0363636363636363e-06, | |
| "loss": 0.9644, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.633204633204633, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 190.3777, | |
| "eval_samples_per_second": 36.885, | |
| "eval_steps_per_second": 4.612, | |
| "eval_wer": 0.5237557295521535, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.633204633204633, | |
| "step": 6000, | |
| "total_flos": 2.5308493485736165e+19, | |
| "train_loss": 0.9557839482625325, | |
| "train_runtime": 12299.0724, | |
| "train_samples_per_second": 15.611, | |
| "train_steps_per_second": 0.488 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.5308493485736165e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |